Raul Davalos Monteiro - Natural exposure and cyclic corrosion testing for the assessment of the performance of powder-coated steel

Presentation - pdf

Raul Davalos-Monteiro a,b, Gianfranco D’Ambrosio b, Xiaorong Zhou a, Simon Gibbon c and Michele Curionia
Corrosion & Protection Centre, School of Materials, The University of Manchester, Manchester, UK

b AkzoNobel, Powder Coatings, Como, Italy
c AkzoNobel R&D, Felling, UK

This work studied the kinetics of corrosion propagation in natural exposure tests of powder coated steel samples and the relationship between natural exposure test results and accelerated cyclic corrosion testing performed according to the ISO-20340 standard. The results from a large number of test panels painted with 26 different powder coating systems suggest that, generally, corrosion propagation in natural exposure is defined by an initial growth in the corrosion area followed by a stage where the corroded area increases very slowly.

Consequently, the relative performance ranking of the coating systems fluctuates considerably as a function of the testing time. Correlations between natural exposure results and accelerated cyclic test results were not found, and for most types of coating classes, a good performance in one test is related with a poor performance in the other.


From  Philip Gill (Cranfield Uni) : @Raul.   Are your observations of differences between natural vs accelerated testing  due to the accelerated test conditions driving unrealistic chemical mechanisms, or does the ISO standard use inaccurate kinetic values for its predictions?  (Normally both in my experience).

Answer: This is a good question. I think the ISO cyclic test that was selected for this project has components that are way more aggressive than the exposure site that we had for the natural exposure test. Of course it has an influence,  of course has thermal stresses impacted from 60C and -20C in the same week, this is one of the limitations of accelerated corrosion test, that you cannot simulate them service performance or the environment, there is always a different response when you compare these two things. In literature some people have found good grade of correlation, so it was worth trying this. If you don’t scribe them they last a long time. I’m trying to combine these results with the microstructure and electrochemistry.

From  Simon Gibbon :  What stops corrosion just covering the whole metal sample over time?
Answer: The coating, I may have misunderstood. The coatings were never intact in the test and that is where corrosion starts.

From  Simon Gibbon : I can see that the defect is where the corrosion starts and so limited corrosion to start with.  This suggests that when you see a tractor outside and all the coatings is gone and corrosion is everywhere, it is a combination of corrosion and other local damage mechanisms to the coating meaning that corrosion starts in lots of different place which all join up.  Rather than one defect and all the corrosion is caused by that single defect.
Answer : Yes I think the degradation of those type of protective powder coatings  is also a combination of corrosion and local damages. I have tested the systems with intact conditions, and if present degradation was only at the edges (where the geometry of the edge influences the formation of weaker coating layers).